aboutsummaryrefslogblamecommitdiffstats
path: root/arch/ppc64/kernel/mf.c
blob: d98bebf7042faf990e63a8732c8dab4c91991033 (plain) (tree)
1
2
3
4
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
811
812
813
814
815
816
817
818
819
820
821
822
823
824
825
826
827
828
829
830
831
832
833
834
835
836
837
838
839
840
841
842
843
844
845
846
847
848
849
850
851
852
853
854
855
856
857
858
859
860
861
862
863
864
865
866
867
868
869
870
871
872
873
874
875
876
877
878
879
880
881
882
883
884
885
886
887
888
889
890
891
892
893
894
895
896
897
898
899
900
901
902
903
904
905
906
907
908
909
910
911
912
913
914
915
916
917
918
919
920
921
922
923
924
925
926
927
928
929
930
931
932
933
934
935
936
937
938
939
940
941
942
943
944
945
946
947
948
949
950
951
952
953
954
955
956
957
958
959
960
961
962
963
964
965
966
967
968
969
970
971
972
973
974
975
976
977
978
979
980
981
982
983
984
985
986
987
988
989
990
991
992
993
994
995
996
997
998
999
1000
1001
1002
1003
1004
1005
1006
1007
1008
1009
1010
1011
1012
1013
1014
1015
1016
1017
1018
1019
1020
1021
1022
1023
1024
1025
1026
1027
1028
1029
1030
1031
1032
1033
1034
1035
1036
1037
1038
1039
1040
1041
1042
1043
1044
1045
1046
1047
1048
1049
1050
1051
1052
1053
1054
1055
1056
1057
1058
1059
1060
1061
1062
1063
1064
1065
1066
1067
1068
1069
1070
1071
1072
1073
1074
1075
1076
1077
1078
1079
1080
1081
1082
1083
1084
1085
1086
1087
1088
1089
1090
1091
1092
1093
1094
1095
1096
1097
1098
1099
1100
1101
1102
1103
1104
1105
1106
1107
1108
1109
1110
1111
1112
1113
1114
1115
1116
1117
1118
1119
1120
1121
1122
1123
1124
1125
1126
1127
1128
1129
1130
1131
1132
1133
1134
1135
1136
1137
1138
1139
1140
1141
1142
1143
1144
1145
1146
1147
1148
1149
1150
1151
1152
1153
1154
1155
1156
1157
1158
1159
1160
1161
1162
1163
1164
1165
1166
1167
1168
1169
1170
1171
1172
1173
1174
1175
1176
1177
1178
1179
1180
1181
1182
1183
1184
1185
1186
1187
1188
1189
1190
1191
1192
1193
1194
1195
1196
1197
1198
1199
1200
1201
1202
1203
1204
1205
1206
1207
1208
1209
1210
1211
1212
1213
1214
1215
1216
1217
1218
1219
1220
1221
1222
1223
1224
1225
1226
1227
1228
1229
1230
1231
1232
1233
1234
1235
1236
1237
1238
1239
1240
1241
1242
1243
1244
1245
1246
1247
1248
1249
1250
1251
1252
1253
1254
1255
1256
1257
1258
1259
1260
1261
1262
1263
1264
1265
1266
1267
1268
1269
1270
1271
1272
1273
1274
1275
1276
1277
1278
1279
1280
1281
1282
1283


                                                         
                                                             

































                                                                             
                     


                                   
                                  















































































































































































































































































































































































































































































































































































































































































                                                                                                                           
                                                              
 


                         
                 





                                 
                          

         

                                  









                                          


























                                    
























































                                                                               
























































































































































































































































































































































































































































































                                                                                         
/*
  * mf.c
  * Copyright (C) 2001 Troy D. Armstrong  IBM Corporation
  * Copyright (C) 2004-2005 Stephen Rothwell  IBM Corporation
  *
  * This modules exists as an interface between a Linux secondary partition
  * running on an iSeries and the primary partition's Virtual Service
  * Processor (VSP) object.  The VSP has final authority over powering on/off
  * all partitions in the iSeries.  It also provides miscellaneous low-level
  * machine facility type operations.
  *
  *
  * This program is free software; you can redistribute it and/or modify
  * it under the terms of the GNU General Public License as published by
  * the Free Software Foundation; either version 2 of the License, or
  * (at your option) any later version.
  *
  * This program is distributed in the hope that it will be useful,
  * but WITHOUT ANY WARRANTY; without even the implied warranty of
  * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
  * GNU General Public License for more details.
  *
  * You should have received a copy of the GNU General Public License
  * along with this program; if not, write to the Free Software
  * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA  02111-1307 USA
  */

#include <linux/types.h>
#include <linux/errno.h>
#include <linux/kernel.h>
#include <linux/init.h>
#include <linux/completion.h>
#include <linux/delay.h>
#include <linux/dma-mapping.h>
#include <linux/bcd.h>

#include <asm/time.h>
#include <asm/uaccess.h>
#include <asm/paca.h>
#include <asm/iSeries/vio.h>
#include <asm/iSeries/mf.h>
#include <asm/iSeries/HvLpConfig.h>
#include <asm/iSeries/ItLpQueue.h>

/*
 * This is the structure layout for the Machine Facilites LPAR event
 * flows.
 */
struct vsp_cmd_data {
	u64 token;
	u16 cmd;
	HvLpIndex lp_index;
	u8 result_code;
	u32 reserved;
	union {
		u64 state;	/* GetStateOut */
		u64 ipl_type;	/* GetIplTypeOut, Function02SelectIplTypeIn */
		u64 ipl_mode;	/* GetIplModeOut, Function02SelectIplModeIn */
		u64 page[4];	/* GetSrcHistoryIn */
		u64 flag;	/* GetAutoIplWhenPrimaryIplsOut,
				   SetAutoIplWhenPrimaryIplsIn,
				   WhiteButtonPowerOffIn,
				   Function08FastPowerOffIn,
				   IsSpcnRackPowerIncompleteOut */
		struct {
			u64 token;
			u64 address_type;
			u64 side;
			u32 length;
			u32 offset;
		} kern;		/* SetKernelImageIn, GetKernelImageIn,
				   SetKernelCmdLineIn, GetKernelCmdLineIn */
		u32 length_out;	/* GetKernelImageOut, GetKernelCmdLineOut */
		u8 reserved[80];
	} sub_data;
};

struct vsp_rsp_data {
	struct completion com;
	struct vsp_cmd_data *response;
};

struct alloc_data {
	u16 size;
	u16 type;
	u32 count;
	u16 reserved1;
	u8 reserved2;
	HvLpIndex target_lp;
};

struct ce_msg_data;

typedef void (*ce_msg_comp_hdlr)(void *token, struct ce_msg_data *vsp_cmd_rsp);

struct ce_msg_comp_data {
	ce_msg_comp_hdlr handler;
	void *token;
};

struct ce_msg_data {
	u8 ce_msg[12];
	char reserved[4];
	struct ce_msg_comp_data *completion;
};

struct io_mf_lp_event {
	struct HvLpEvent hp_lp_event;
	u16 subtype_result_code;
	u16 reserved1;
	u32 reserved2;
	union {
		struct alloc_data alloc;
		struct ce_msg_data ce_msg;
		struct vsp_cmd_data vsp_cmd;
	} data;
};

#define subtype_data(a, b, c, d)	\
		(((a) << 24) + ((b) << 16) + ((c) << 8) + (d))

/*
 * All outgoing event traffic is kept on a FIFO queue.  The first
 * pointer points to the one that is outstanding, and all new
 * requests get stuck on the end.  Also, we keep a certain number of
 * preallocated pending events so that we can operate very early in
 * the boot up sequence (before kmalloc is ready).
 */
struct pending_event {
	struct pending_event *next;
	struct io_mf_lp_event event;
	MFCompleteHandler hdlr;
	char dma_data[72];
	unsigned dma_data_length;
	unsigned remote_address;
};
static spinlock_t pending_event_spinlock;
static struct pending_event *pending_event_head;
static struct pending_event *pending_event_tail;
static struct pending_event *pending_event_avail;
static struct pending_event pending_event_prealloc[16];

/*
 * Put a pending event onto the available queue, so it can get reused.
 * Attention! You must have the pending_event_spinlock before calling!
 */
static void free_pending_event(struct pending_event *ev)
{
	if (ev != NULL) {
		ev->next = pending_event_avail;
		pending_event_avail = ev;
	}
}

/*
 * Enqueue the outbound event onto the stack.  If the queue was
 * empty to begin with, we must also issue it via the Hypervisor
 * interface.  There is a section of code below that will touch
 * the first stack pointer without the protection of the pending_event_spinlock.
 * This is OK, because we know that nobody else will be modifying
 * the first pointer when we do this.
 */
static int signal_event(struct pending_event *ev)
{
	int rc = 0;
	unsigned long flags;
	int go = 1;
	struct pending_event *ev1;
	HvLpEvent_Rc hv_rc;

	/* enqueue the event */
	if (ev != NULL) {
		ev->next = NULL;
		spin_lock_irqsave(&pending_event_spinlock, flags);
		if (pending_event_head == NULL)
			pending_event_head = ev;
		else {
			go = 0;
			pending_event_tail->next = ev;
		}
		pending_event_tail = ev;
		spin_unlock_irqrestore(&pending_event_spinlock, flags);
	}

	/* send the event */
	while (go) {
		go = 0;

		/* any DMA data to send beforehand? */
		if (pending_event_head->dma_data_length > 0)
			HvCallEvent_dmaToSp(pending_event_head->dma_data,
					pending_event_head->remote_address,
					pending_event_head->dma_data_length,
					HvLpDma_Direction_LocalToRemote);

		hv_rc = HvCallEvent_signalLpEvent(
				&pending_event_head->event.hp_lp_event);
		if (hv_rc != HvLpEvent_Rc_Good) {
			printk(KERN_ERR "mf.c: HvCallEvent_signalLpEvent() "
					"failed with %d\n", (int)hv_rc);

			spin_lock_irqsave(&pending_event_spinlock, flags);
			ev1 = pending_event_head;
			pending_event_head = pending_event_head->next;
			if (pending_event_head != NULL)
				go = 1;
			spin_unlock_irqrestore(&pending_event_spinlock, flags);

			if (ev1 == ev)
				rc = -EIO;
			else if (ev1->hdlr != NULL)
				(*ev1->hdlr)((void *)ev1->event.hp_lp_event.xCorrelationToken, -EIO);

			spin_lock_irqsave(&pending_event_spinlock, flags);
			free_pending_event(ev1);
			spin_unlock_irqrestore(&pending_event_spinlock, flags);
		}
	}

	return rc;
}

/*
 * Allocate a new pending_event structure, and initialize it.
 */
static struct pending_event *new_pending_event(void)
{
	struct pending_event *ev = NULL;
	HvLpIndex primary_lp = HvLpConfig_getPrimaryLpIndex();
	unsigned long flags;
	struct HvLpEvent *hev;

	spin_lock_irqsave(&pending_event_spinlock, flags);
	if (pending_event_avail != NULL) {
		ev = pending_event_avail;
		pending_event_avail = pending_event_avail->next;
	}
	spin_unlock_irqrestore(&pending_event_spinlock, flags);
	if (ev == NULL) {
		ev = kmalloc(sizeof(struct pending_event), GFP_ATOMIC);
		if (ev == NULL) {
			printk(KERN_ERR "mf.c: unable to kmalloc %ld bytes\n",
					sizeof(struct pending_event));
			return NULL;
		}
	}
	memset(ev, 0, sizeof(struct pending_event));
	hev = &ev->event.hp_lp_event;
	hev->xFlags.xValid = 1;
	hev->xFlags.xAckType = HvLpEvent_AckType_ImmediateAck;
	hev->xFlags.xAckInd = HvLpEvent_AckInd_DoAck;
	hev->xFlags.xFunction = HvLpEvent_Function_Int;
	hev->xType = HvLpEvent_Type_MachineFac;
	hev->xSourceLp = HvLpConfig_getLpIndex();
	hev->xTargetLp = primary_lp;
	hev->xSizeMinus1 = sizeof(ev->event) - 1;
	hev->xRc = HvLpEvent_Rc_Good;
	hev->xSourceInstanceId = HvCallEvent_getSourceLpInstanceId(primary_lp,
			HvLpEvent_Type_MachineFac);
	hev->xTargetInstanceId = HvCallEvent_getTargetLpInstanceId(primary_lp,
			HvLpEvent_Type_MachineFac);

	return ev;
}

static int signal_vsp_instruction(struct vsp_cmd_data *vsp_cmd)
{
	struct pending_event *ev = new_pending_event();
	int rc;
	struct vsp_rsp_data response;

	if (ev == NULL)
		return -ENOMEM;

	init_completion(&response.com);
	response.response = vsp_cmd;
	ev->event.hp_lp_event.xSubtype = 6;
	ev->event.hp_lp_event.x.xSubtypeData =
		subtype_data('M', 'F',  'V',  'I');
	ev->event.data.vsp_cmd.token = (u64)&response;
	ev->event.data.vsp_cmd.cmd = vsp_cmd->cmd;
	ev->event.data.vsp_cmd.lp_index = HvLpConfig_getLpIndex();
	ev->event.data.vsp_cmd.result_code = 0xFF;
	ev->event.data.vsp_cmd.reserved = 0;
	memcpy(&(ev->event.data.vsp_cmd.sub_data),
			&(vsp_cmd->sub_data), sizeof(vsp_cmd->sub_data));
	mb();

	rc = signal_event(ev);
	if (rc == 0)
		wait_for_completion(&response.com);
	return rc;
}


/*
 * Send a 12-byte CE message to the primary partition VSP object
 */
static int signal_ce_msg(char *ce_msg, struct ce_msg_comp_data *completion)
{
	struct pending_event *ev = new_pending_event();

	if (ev == NULL)
		return -ENOMEM;

	ev->event.hp_lp_event.xSubtype = 0;
	ev->event.hp_lp_event.x.xSubtypeData =
		subtype_data('M',  'F',  'C',  'E');
	memcpy(ev->event.data.ce_msg.ce_msg, ce_msg, 12);
	ev->event.data.ce_msg.completion = completion;
	return signal_event(ev);
}

/*
 * Send a 12-byte CE message (with no data) to the primary partition VSP object
 */
static int signal_ce_msg_simple(u8 ce_op, struct ce_msg_comp_data *completion)
{
	u8 ce_msg[12];

	memset(ce_msg, 0, sizeof(ce_msg));
	ce_msg[3] = ce_op;
	return signal_ce_msg(ce_msg, completion);
}

/*
 * Send a 12-byte CE message and DMA data to the primary partition VSP object
 */
static int dma_and_signal_ce_msg(char *ce_msg,
		struct ce_msg_comp_data *completion, void *dma_data,
		unsigned dma_data_length, unsigned remote_address)
{
	struct pending_event *ev = new_pending_event();

	if (ev == NULL)
		return -ENOMEM;

	ev->event.hp_lp_event.xSubtype = 0;
	ev->event.hp_lp_event.x.xSubtypeData =
		subtype_data('M', 'F', 'C', 'E');
	memcpy(ev->event.data.ce_msg.ce_msg, ce_msg, 12);
	ev->event.data.ce_msg.completion = completion;
	memcpy(ev->dma_data, dma_data, dma_data_length);
	ev->dma_data_length = dma_data_length;
	ev->remote_address = remote_address;
	return signal_event(ev);
}

/*
 * Initiate a nice (hopefully) shutdown of Linux.  We simply are
 * going to try and send the init process a SIGINT signal.  If
 * this fails (why?), we'll simply force it off in a not-so-nice
 * manner.
 */
static int shutdown(void)
{
	int rc = kill_proc(1, SIGINT, 1);

	if (rc) {
		printk(KERN_ALERT "mf.c: SIGINT to init failed (%d), "
				"hard shutdown commencing\n", rc);
		mf_power_off();
	} else
		printk(KERN_INFO "mf.c: init has been successfully notified "
				"to proceed with shutdown\n");
	return rc;
}

/*
 * The primary partition VSP object is sending us a new
 * event flow.  Handle it...
 */
static void handle_int(struct io_mf_lp_event *event)
{
	struct ce_msg_data *ce_msg_data;
	struct ce_msg_data *pce_msg_data;
	unsigned long flags;
	struct pending_event *pev;

	/* ack the interrupt */
	event->hp_lp_event.xRc = HvLpEvent_Rc_Good;
	HvCallEvent_ackLpEvent(&event->hp_lp_event);

	/* process interrupt */
	switch (event->hp_lp_event.xSubtype) {
	case 0:	/* CE message */
		ce_msg_data = &event->data.ce_msg;
		switch (ce_msg_data->ce_msg[3]) {
		case 0x5B:	/* power control notification */
			if ((ce_msg_data->ce_msg[5] & 0x20) != 0) {
				printk(KERN_INFO "mf.c: Commencing partition shutdown\n");
				if (shutdown() == 0)
					signal_ce_msg_simple(0xDB, NULL);
			}
			break;
		case 0xC0:	/* get time */
			spin_lock_irqsave(&pending_event_spinlock, flags);
			pev = pending_event_head;
			if (pev != NULL)
				pending_event_head = pending_event_head->next;
			spin_unlock_irqrestore(&pending_event_spinlock, flags);
			if (pev == NULL)
				break;
			pce_msg_data = &pev->event.data.ce_msg;
			if (pce_msg_data->ce_msg[3] != 0x40)
				break;
			if (pce_msg_data->completion != NULL) {
				ce_msg_comp_hdlr handler =
					pce_msg_data->completion->handler;
				void *token = pce_msg_data->completion->token;

				if (handler != NULL)
					(*handler)(token, ce_msg_data);
			}
			spin_lock_irqsave(&pending_event_spinlock, flags);
			free_pending_event(pev);
			spin_unlock_irqrestore(&pending_event_spinlock, flags);
			/* send next waiting event */
			if (pending_event_head != NULL)
				signal_event(NULL);
			break;
		}
		break;
	case 1:	/* IT sys shutdown */
		printk(KERN_INFO "mf.c: Commencing system shutdown\n");
		shutdown();
		break;
	}
}

/*
 * The primary partition VSP object is acknowledging the receipt
 * of a flow we sent to them.  If there are other flows queued
 * up, we must send another one now...
 */
static void handle_ack(struct io_mf_lp_event *event)
{
	unsigned long flags;
	struct pending_event *two = NULL;
	unsigned long free_it = 0;
	struct ce_msg_data *ce_msg_data;
	struct ce_msg_data *pce_msg_data;
	struct vsp_rsp_data *rsp;

	/* handle current event */
	if (pending_event_head == NULL) {
		printk(KERN_ERR "mf.c: stack empty for receiving ack\n");
		return;
	}

	switch (event->hp_lp_event.xSubtype) {
	case 0:     /* CE msg */
		ce_msg_data = &event->data.ce_msg;
		if (ce_msg_data->ce_msg[3] != 0x40) {
			free_it = 1;
			break;
		}
		if (ce_msg_data->ce_msg[2] == 0)
			break;
		free_it = 1;
		pce_msg_data = &pending_event_head->event.data.ce_msg;
		if (pce_msg_data->completion != NULL) {
			ce_msg_comp_hdlr handler =
				pce_msg_data->completion->handler;
			void *token = pce_msg_data->completion->token;

			if (handler != NULL)
				(*handler)(token, ce_msg_data);
		}
		break;
	case 4:	/* allocate */
	case 5:	/* deallocate */
		if (pending_event_head->hdlr != NULL)
			(*pending_event_head->hdlr)((void *)event->hp_lp_event.xCorrelationToken, event->data.alloc.count);
		free_it = 1;
		break;
	case 6:
		free_it = 1;
		rsp = (struct vsp_rsp_data *)event->data.vsp_cmd.token;
		if (rsp == NULL) {
			printk(KERN_ERR "mf.c: no rsp\n");
			break;
		}
		if (rsp->response != NULL)
			memcpy(rsp->response, &event->data.vsp_cmd,
					sizeof(event->data.vsp_cmd));
		complete(&rsp->com);
		break;
	}

	/* remove from queue */
	spin_lock_irqsave(&pending_event_spinlock, flags);
	if ((pending_event_head != NULL) && (free_it == 1)) {
		struct pending_event *oldHead = pending_event_head;

		pending_event_head = pending_event_head->next;
		two = pending_event_head;
		free_pending_event(oldHead);
	}
	spin_unlock_irqrestore(&pending_event_spinlock, flags);

	/* send next waiting event */
	if (two != NULL)
		signal_event(NULL);
}

/*
 * This is the generic event handler we are registering with
 * the Hypervisor.  Ensure the flows are for us, and then
 * parse it enough to know if it is an interrupt or an
 * acknowledge.
 */
static void hv_handler(struct HvLpEvent *event, struct pt_regs *regs)
{
	if ((event != NULL) && (event->xType == HvLpEvent_Type_MachineFac)) {
		switch(event->xFlags.xFunction) {
		case HvLpEvent_Function_Ack:
			handle_ack((struct io_mf_lp_event *)event);
			break;
		case HvLpEvent_Function_Int:
			handle_int((struct io_mf_lp_event *)event);
			break;
		default:
			printk(KERN_ERR "mf.c: non ack/int event received\n");
			break;
		}
	} else
		printk(KERN_ERR "mf.c: alien event received\n");
}

/*
 * Global kernel interface to allocate and seed events into the
 * Hypervisor.
 */
void mf_allocate_lp_events(HvLpIndex target_lp, HvLpEvent_Type type,
		unsigned size, unsigned count, MFCompleteHandler hdlr,
		void *user_token)
{
	struct pending_event *ev = new_pending_event();
	int rc;

	if (ev == NULL) {
		rc = -ENOMEM;
	} else {
		ev->event.hp_lp_event.xSubtype = 4;
		ev->event.hp_lp_event.xCorrelationToken = (u64)user_token;
		ev->event.hp_lp_event.x.xSubtypeData =
			subtype_data('M', 'F', 'M', 'A');
		ev->event.data.alloc.target_lp = target_lp;
		ev->event.data.alloc.type = type;
		ev->event.data.alloc.size = size;
		ev->event.data.alloc.count = count;
		ev->hdlr = hdlr;
		rc = signal_event(ev);
	}
	if ((rc != 0) && (hdlr != NULL))
		(*hdlr)(user_token, rc);
}
EXPORT_SYMBOL(mf_allocate_lp_events);

/*
 * Global kernel interface to unseed and deallocate events already in
 * Hypervisor.
 */
void mf_deallocate_lp_events(HvLpIndex target_lp, HvLpEvent_Type type,
		unsigned count, MFCompleteHandler hdlr, void *user_token)
{
	struct pending_event *ev = new_pending_event();
	int rc;

	if (ev == NULL)
		rc = -ENOMEM;
	else {
		ev->event.hp_lp_event.xSubtype = 5;
		ev->event.hp_lp_event.xCorrelationToken = (u64)user_token;
		ev->event.hp_lp_event.x.xSubtypeData =
			subtype_data('M', 'F', 'M', 'D');
		ev->event.data.alloc.target_lp = target_lp;
		ev->event.data.alloc.type = type;
		ev->event.data.alloc.count = count;
		ev->hdlr = hdlr;
		rc = signal_event(ev);
	}
	if ((rc != 0) && (hdlr != NULL))
		(*hdlr)(user_token, rc);
}
EXPORT_SYMBOL(mf_deallocate_lp_events);

/*
 * Global kernel interface to tell the VSP object in the primary
 * partition to power this partition off.
 */
void mf_power_off(void)
{
	printk(KERN_INFO "mf.c: Down it goes...\n");
	signal_ce_msg_simple(0x4d, NULL);
	for (;;)
		;
}

/*
 * Global kernel interface to tell the VSP object in the primary
 * partition to reboot this partition.
 */
void mf_reboot(void)
{
	printk(KERN_INFO "mf.c: Preparing to bounce...\n");
	signal_ce_msg_simple(0x4e, NULL);
	for (;;)
		;
}

/*
 * Display a single word SRC onto the VSP control panel.
 */
void mf_display_src(u32 word)
{
	u8 ce[12];

	memset(ce, 0, sizeof(ce));
	ce[3] = 0x4a;
	ce[7] = 0x01;
	ce[8] = word >> 24;
	ce[9] = word >> 16;
	ce[10] = word >> 8;
	ce[11] = word;
	signal_ce_msg(ce, NULL);
}

/*
 * Display a single word SRC of the form "PROGXXXX" on the VSP control panel.
 */
void mf_display_progress(u16 value)
{
	u8 ce[12];
	u8 src[72];

	memcpy(ce, "\x00\x00\x04\x4A\x00\x00\x00\x48\x00\x00\x00\x00", 12);
	memcpy(src, "\x01\x00\x00\x01\x00\x00\x00\x00\x00\x00\x00\x00"
		"\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00"
		"\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00"
		"\x00\x00\x00\x00PROGxxxx                        ",
		72);
	src[6] = value >> 8;
	src[7] = value & 255;
	src[44] = "0123456789ABCDEF"[(value >> 12) & 15];
	src[45] = "0123456789ABCDEF"[(value >> 8) & 15];
	src[46] = "0123456789ABCDEF"[(value >> 4) & 15];
	src[47] = "0123456789ABCDEF"[value & 15];
	dma_and_signal_ce_msg(ce, NULL, src, sizeof(src), 9 * 64 * 1024);
}

/*
 * Clear the VSP control panel.  Used to "erase" an SRC that was
 * previously displayed.
 */
void mf_clear_src(void)
{
	signal_ce_msg_simple(0x4b, NULL);
}

/*
 * Initialization code here.
 */
void mf_init(void)
{
	int i;

	/* initialize */
	spin_lock_init(&pending_event_spinlock);
	for (i = 0;
	     i < sizeof(pending_event_prealloc) / sizeof(*pending_event_prealloc);
	     ++i)
		free_pending_event(&pending_event_prealloc[i]);
	HvLpEvent_registerHandler(HvLpEvent_Type_MachineFac, &hv_handler);

	/* virtual continue ack */
	signal_ce_msg_simple(0x57, NULL);

	/* initialization complete */
	printk(KERN_NOTICE "mf.c: iSeries Linux LPAR Machine Facilities "
			"initialized\n");
}

struct rtc_time_data {
	struct completion com;
	struct ce_msg_data ce_msg;
	int rc;
};

static void get_rtc_time_complete(void *token, struct ce_msg_data *ce_msg)
{
	struct rtc_time_data *rtc = token;

	memcpy(&rtc->ce_msg, ce_msg, sizeof(rtc->ce_msg));
	rtc->rc = 0;
	complete(&rtc->com);
}

static int rtc_set_tm(int rc, u8 *ce_msg, struct rtc_time *tm)
{
	tm->tm_wday = 0;
	tm->tm_yday = 0;
	tm->tm_isdst = 0;
	if (rc) {
		tm->tm_sec = 0;
		tm->tm_min = 0;
		tm->tm_hour = 0;
		tm->tm_mday = 15;
		tm->tm_mon = 5;
		tm->tm_year = 52;
		return rc;
	}

	if ((ce_msg[2] == 0xa9) ||
	    (ce_msg[2] == 0xaf)) {
		/* TOD clock is not set */
		tm->tm_sec = 1;
		tm->tm_min = 1;
		tm->tm_hour = 1;
		tm->tm_mday = 10;
		tm->tm_mon = 8;
		tm->tm_year = 71;
		mf_set_rtc(tm);
	}
	{
		u8 year = ce_msg[5];
		u8 sec = ce_msg[6];
		u8 min = ce_msg[7];
		u8 hour = ce_msg[8];
		u8 day = ce_msg[10];
		u8 mon = ce_msg[11];

		BCD_TO_BIN(sec);
		BCD_TO_BIN(min);
		BCD_TO_BIN(hour);
		BCD_TO_BIN(day);
		BCD_TO_BIN(mon);
		BCD_TO_BIN(year);

		if (year <= 69)
			year += 100;

		tm->tm_sec = sec;
		tm->tm_min = min;
		tm->tm_hour = hour;
		tm->tm_mday = day;
		tm->tm_mon = mon;
		tm->tm_year = year;
	}

	return 0;
}

int mf_get_rtc(struct rtc_time *tm)
{
	struct ce_msg_comp_data ce_complete;
	struct rtc_time_data rtc_data;
	int rc;

	memset(&ce_complete, 0, sizeof(ce_complete));
	memset(&rtc_data, 0, sizeof(rtc_data));
	init_completion(&rtc_data.com);
	ce_complete.handler = &get_rtc_time_complete;
	ce_complete.token = &rtc_data;
	rc = signal_ce_msg_simple(0x40, &ce_complete);
	if (rc)
		return rc;
	wait_for_completion(&rtc_data.com);
	return rtc_set_tm(rtc_data.rc, rtc_data.ce_msg.ce_msg, tm);
}

struct boot_rtc_time_data {
	int busy;
	struct ce_msg_data ce_msg;
	int rc;
};

static void get_boot_rtc_time_complete(void *token, struct ce_msg_data *ce_msg)
{
	struct boot_rtc_time_data *rtc = token;

	memcpy(&rtc->ce_msg, ce_msg, sizeof(rtc->ce_msg));
	rtc->rc = 0;
	rtc->busy = 0;
}

int mf_get_boot_rtc(struct rtc_time *tm)
{
	struct ce_msg_comp_data ce_complete;
	struct boot_rtc_time_data rtc_data;
	int rc;

	memset(&ce_complete, 0, sizeof(ce_complete));
	memset(&rtc_data, 0, sizeof(rtc_data));
	rtc_data.busy = 1;
	ce_complete.handler = &get_boot_rtc_time_complete;
	ce_complete.token = &rtc_data;
	rc = signal_ce_msg_simple(0x40, &ce_complete);
	if (rc)
		return rc;
	/* We need to poll here as we are not yet taking interrupts */
	while (rtc_data.busy) {
		extern unsigned long lpevent_count;
		struct ItLpQueue *lpq = get_paca()->lpqueue_ptr;
		if (lpq && ItLpQueue_isLpIntPending(lpq))
			lpevent_count += ItLpQueue_process(lpq, NULL);
	}
	return rtc_set_tm(rtc_data.rc, rtc_data.ce_msg.ce_msg, tm);
}

int mf_set_rtc(struct rtc_time *tm)
{
	char ce_time[12];
	u8 day, mon, hour, min, sec, y1, y2;
	unsigned year;

	year = 1900 + tm->tm_year;
	y1 = year / 100;
	y2 = year % 100;

	sec = tm->tm_sec;
	min = tm->tm_min;
	hour = tm->tm_hour;
	day = tm->tm_mday;
	mon = tm->tm_mon + 1;

	BIN_TO_BCD(sec);
	BIN_TO_BCD(min);
	BIN_TO_BCD(hour);
	BIN_TO_BCD(mon);
	BIN_TO_BCD(day);
	BIN_TO_BCD(y1);
	BIN_TO_BCD(y2);

	memset(ce_time, 0, sizeof(ce_time));
	ce_time[3] = 0x41;
	ce_time[4] = y1;
	ce_time[5] = y2;
	ce_time[6] = sec;
	ce_time[7] = min;
	ce_time[8] = hour;
	ce_time[10] = day;
	ce_time[11] = mon;

	return signal_ce_msg(ce_time, NULL);
}

#ifdef CONFIG_PROC_FS

static int proc_mf_dump_cmdline(char *page, char **start, off_t off,
		int count, int *eof, void *data)
{
	int len;
	char *p;
	struct vsp_cmd_data vsp_cmd;
	int rc;
	dma_addr_t dma_addr;

	/* The HV appears to return no more than 256 bytes of command line */
	if (off >= 256)
		return 0;
	if ((off + count) > 256)
		count = 256 - off;

	dma_addr = dma_map_single(iSeries_vio_dev, page, off + count,
			DMA_FROM_DEVICE);
	if (dma_mapping_error(dma_addr))
		return -ENOMEM;
	memset(page, 0, off + count);
	memset(&vsp_cmd, 0, sizeof(vsp_cmd));
	vsp_cmd.cmd = 33;
	vsp_cmd.sub_data.kern.token = dma_addr;
	vsp_cmd.sub_data.kern.address_type = HvLpDma_AddressType_TceIndex;
	vsp_cmd.sub_data.kern.side = (u64)data;
	vsp_cmd.sub_data.kern.length = off + count;
	mb();
	rc = signal_vsp_instruction(&vsp_cmd);
	dma_unmap_single(iSeries_vio_dev, dma_addr, off + count,
			DMA_FROM_DEVICE);
	if (rc)
		return rc;
	if (vsp_cmd.result_code != 0)
		return -ENOMEM;
	p = page;
	len = 0;
	while (len < (off + count)) {
		if ((*p == '\0') || (*p == '\n')) {
			if (*p == '\0')
				*p = '\n';
			p++;
			len++;
			*eof = 1;
			break;
		}
		p++;
		len++;
	}

	if (len < off) {
		*eof = 1;
		len = 0;
	}
	return len;
}

#if 0
static int mf_getVmlinuxChunk(char *buffer, int *size, int offset, u64 side)
{
	struct vsp_cmd_data vsp_cmd;
	int rc;
	int len = *size;
	dma_addr_t dma_addr;

	dma_addr = dma_map_single(iSeries_vio_dev, buffer, len,
			DMA_FROM_DEVICE);
	memset(buffer, 0, len);
	memset(&vsp_cmd, 0, sizeof(vsp_cmd));
	vsp_cmd.cmd = 32;
	vsp_cmd.sub_data.kern.token = dma_addr;
	vsp_cmd.sub_data.kern.address_type = HvLpDma_AddressType_TceIndex;
	vsp_cmd.sub_data.kern.side = side;
	vsp_cmd.sub_data.kern.offset = offset;
	vsp_cmd.sub_data.kern.length = len;
	mb();
	rc = signal_vsp_instruction(&vsp_cmd);
	if (rc == 0) {
		if (vsp_cmd.result_code == 0)
			*size = vsp_cmd.sub_data.length_out;
		else
			rc = -ENOMEM;
	}

	dma_unmap_single(iSeries_vio_dev, dma_addr, len, DMA_FROM_DEVICE);

	return rc;
}

static int proc_mf_dump_vmlinux(char *page, char **start, off_t off,
		int count, int *eof, void *data)
{
	int sizeToGet = count;

	if (!capable(CAP_SYS_ADMIN))
		return -EACCES;

	if (mf_getVmlinuxChunk(page, &sizeToGet, off, (u64)data) == 0) {
		if (sizeToGet != 0) {
			*start = page + off;
			return sizeToGet;
		}
		*eof = 1;
		return 0;
	}
	*eof = 1;
	return 0;
}
#endif

static int proc_mf_dump_side(char *page, char **start, off_t off,
		int count, int *eof, void *data)
{
	int len;
	char mf_current_side = ' ';
	struct vsp_cmd_data vsp_cmd;

	memset(&vsp_cmd, 0, sizeof(vsp_cmd));
	vsp_cmd.cmd = 2;
	vsp_cmd.sub_data.ipl_type = 0;
	mb();

	if (signal_vsp_instruction(&vsp_cmd) == 0) {
		if (vsp_cmd.result_code == 0) {
			switch (vsp_cmd.sub_data.ipl_type) {
			case 0:	mf_current_side = 'A';
				break;
			case 1:	mf_current_side = 'B';
				break;
			case 2:	mf_current_side = 'C';
				break;
			default:	mf_current_side = 'D';
				break;
			}
		}
	}

	len = sprintf(page, "%c\n", mf_current_side);

	if (len <= (off + count))
		*eof = 1;
	*start = page + off;
	len -= off;
	if (len > count)
		len = count;
	if (len < 0)
		len = 0;
	return len;
}

static int proc_mf_change_side(struct file *file, const char __user *buffer,
		unsigned long count, void *data)
{
	char side;
	u64 newSide;
	struct vsp_cmd_data vsp_cmd;

	if (!capable(CAP_SYS_ADMIN))
		return -EACCES;

	if (count == 0)
		return 0;

	if (get_user(side, buffer))
		return -EFAULT;

	switch (side) {
	case 'A':	newSide = 0;
			break;
	case 'B':	newSide = 1;
			break;
	case 'C':	newSide = 2;
			break;
	case 'D':	newSide = 3;
			break;
	default:
		printk(KERN_ERR "mf_proc.c: proc_mf_change_side: invalid side\n");
		return -EINVAL;
	}

	memset(&vsp_cmd, 0, sizeof(vsp_cmd));
	vsp_cmd.sub_data.ipl_type = newSide;
	vsp_cmd.cmd = 10;

	(void)signal_vsp_instruction(&vsp_cmd);

	return count;
}

#if 0
static void mf_getSrcHistory(char *buffer, int size)
{
	struct IplTypeReturnStuff return_stuff;
	struct pending_event *ev = new_pending_event();
	int rc = 0;
	char *pages[4];

	pages[0] = kmalloc(4096, GFP_ATOMIC);
	pages[1] = kmalloc(4096, GFP_ATOMIC);
	pages[2] = kmalloc(4096, GFP_ATOMIC);
	pages[3] = kmalloc(4096, GFP_ATOMIC);
	if ((ev == NULL) || (pages[0] == NULL) || (pages[1] == NULL)
			 || (pages[2] == NULL) || (pages[3] == NULL))
		return -ENOMEM;

	return_stuff.xType = 0;
	return_stuff.xRc = 0;
	return_stuff.xDone = 0;
	ev->event.hp_lp_event.xSubtype = 6;
	ev->event.hp_lp_event.x.xSubtypeData =
		subtype_data('M', 'F', 'V', 'I');
	ev->event.data.vsp_cmd.xEvent = &return_stuff;
	ev->event.data.vsp_cmd.cmd = 4;
	ev->event.data.vsp_cmd.lp_index = HvLpConfig_getLpIndex();
	ev->event.data.vsp_cmd.result_code = 0xFF;
	ev->event.data.vsp_cmd.reserved = 0;
	ev->event.data.vsp_cmd.sub_data.page[0] = ISERIES_HV_ADDR(pages[0]);
	ev->event.data.vsp_cmd.sub_data.page[1] = ISERIES_HV_ADDR(pages[1]);
	ev->event.data.vsp_cmd.sub_data.page[2] = ISERIES_HV_ADDR(pages[2]);
	ev->event.data.vsp_cmd.sub_data.page[3] = ISERIES_HV_ADDR(pages[3]);
	mb();
	if (signal_event(ev) != 0)
		return;

 	while (return_stuff.xDone != 1)
 		udelay(10);
 	if (return_stuff.xRc == 0)
 		memcpy(buffer, pages[0], size);
	kfree(pages[0]);
	kfree(pages[1]);
	kfree(pages[2]);
	kfree(pages[3]);
}
#endif

static int proc_mf_dump_src(char *page, char **start, off_t off,
		int count, int *eof, void *data)
{
#if 0
	int len;

	mf_getSrcHistory(page, count);
	len = count;
	len -= off;
	if (len < count) {
		*eof = 1;
		if (len <= 0)
			return 0;
	} else
		len = count;
	*start = page + off;
	return len;
#else
	return 0;
#endif
}

static int proc_mf_change_src(struct file *file, const char __user *buffer,
		unsigned long count, void *data)
{
	char stkbuf[10];

	if (!capable(CAP_SYS_ADMIN))
		return -EACCES;

	if ((count < 4) && (count != 1)) {
		printk(KERN_ERR "mf_proc: invalid src\n");
		return -EINVAL;
	}

	if (count > (sizeof(stkbuf) - 1))
		count = sizeof(stkbuf) - 1;
	if (copy_from_user(stkbuf, buffer, count))
		return -EFAULT;

	if ((count == 1) && (*stkbuf == '\0'))
		mf_clear_src();
	else
		mf_display_src(*(u32 *)stkbuf);

	return count;
}

static int proc_mf_change_cmdline(struct file *file, const char __user *buffer,
		unsigned long count, void *data)
{
	struct vsp_cmd_data vsp_cmd;
	dma_addr_t dma_addr;
	char *page;
	int ret = -EACCES;

	if (!capable(CAP_SYS_ADMIN))
		goto out;

	dma_addr = 0;
	page = dma_alloc_coherent(iSeries_vio_dev, count, &dma_addr,
			GFP_ATOMIC);
	ret = -ENOMEM;
	if (page == NULL)
		goto out;

	ret = -EFAULT;
	if (copy_from_user(page, buffer, count))
		goto out_free;

	memset(&vsp_cmd, 0, sizeof(vsp_cmd));
	vsp_cmd.cmd = 31;
	vsp_cmd.sub_data.kern.token = dma_addr;
	vsp_cmd.sub_data.kern.address_type = HvLpDma_AddressType_TceIndex;
	vsp_cmd.sub_data.kern.side = (u64)data;
	vsp_cmd.sub_data.kern.length = count;
	mb();
	(void)signal_vsp_instruction(&vsp_cmd);
	ret = count;

out_free:
	dma_free_coherent(iSeries_vio_dev, count, page, dma_addr);
out:
	return ret;
}

static ssize_t proc_mf_change_vmlinux(struct file *file,
				      const char __user *buf,
				      size_t count, loff_t *ppos)
{
	struct proc_dir_entry *dp = PDE(file->f_dentry->d_inode);
	ssize_t rc;
	dma_addr_t dma_addr;
	char *page;
	struct vsp_cmd_data vsp_cmd;

	rc = -EACCES;
	if (!capable(CAP_SYS_ADMIN))
		goto out;

	dma_addr = 0;
	page = dma_alloc_coherent(iSeries_vio_dev, count, &dma_addr,
			GFP_ATOMIC);
	rc = -ENOMEM;
	if (page == NULL) {
		printk(KERN_ERR "mf.c: couldn't allocate memory to set vmlinux chunk\n");
		goto out;
	}
	rc = -EFAULT;
	if (copy_from_user(page, buf, count))
		goto out_free;

	memset(&vsp_cmd, 0, sizeof(vsp_cmd));
	vsp_cmd.cmd = 30;
	vsp_cmd.sub_data.kern.token = dma_addr;
	vsp_cmd.sub_data.kern.address_type = HvLpDma_AddressType_TceIndex;
	vsp_cmd.sub_data.kern.side = (u64)dp->data;
	vsp_cmd.sub_data.kern.offset = *ppos;
	vsp_cmd.sub_data.kern.length = count;
	mb();
	rc = signal_vsp_instruction(&vsp_cmd);
	if (rc)
		goto out_free;
	rc = -ENOMEM;
	if (vsp_cmd.result_code != 0)
		goto out_free;

	*ppos += count;
	rc = count;
out_free:
	dma_free_coherent(iSeries_vio_dev, count, page, dma_addr);
out:
	return rc;
}

static struct file_operations proc_vmlinux_operations = {
	.write		= proc_mf_change_vmlinux,
};

static int __init mf_proc_init(void)
{
	struct proc_dir_entry *mf_proc_root;
	struct proc_dir_entry *ent;
	struct proc_dir_entry *mf;
	char name[2];
	int i;

	mf_proc_root = proc_mkdir("iSeries/mf", NULL);
	if (!mf_proc_root)
		return 1;

	name[1] = '\0';
	for (i = 0; i < 4; i++) {
		name[0] = 'A' + i;
		mf = proc_mkdir(name, mf_proc_root);
		if (!mf)
			return 1;

		ent = create_proc_entry("cmdline", S_IFREG|S_IRUSR|S_IWUSR, mf);
		if (!ent)
			return 1;
		ent->nlink = 1;
		ent->data = (void *)(long)i;
		ent->read_proc = proc_mf_dump_cmdline;
		ent->write_proc = proc_mf_change_cmdline;

		if (i == 3)	/* no vmlinux entry for 'D' */
			continue;

		ent = create_proc_entry("vmlinux", S_IFREG|S_IWUSR, mf);
		if (!ent)
			return 1;
		ent->nlink = 1;
		ent->data = (void *)(long)i;
		ent->proc_fops = &proc_vmlinux_operations;
	}

	ent = create_proc_entry("side", S_IFREG|S_IRUSR|S_IWUSR, mf_proc_root);
	if (!ent)
		return 1;
	ent->nlink = 1;
	ent->data = (void *)0;
	ent->read_proc = proc_mf_dump_side;
	ent->write_proc = proc_mf_change_side;

	ent = create_proc_entry("src", S_IFREG|S_IRUSR|S_IWUSR, mf_proc_root);
	if (!ent)
		return 1;
	ent->nlink = 1;
	ent->data = (void *)0;
	ent->read_proc = proc_mf_dump_src;
	ent->write_proc = proc_mf_change_src;

	return 0;
}

__initcall(mf_proc_init);

#endif /* CONFIG_PROC_FS */